The Pause that Refreshes

How will the Fed respond to the recent drops in the stock market? We might get a hint from the Fed’s conduct after last August’s turbulence, which now looks like a dress rehearsal for the current problems. In that event, the Fed chose to forgo the widely-signaled liftoff in September, but then implemented liftoff amid quieter markets in December. The lesson? In my view, the main lesson is that market distress may lead to a pause in policy plans, but unless macro fundamentals soon echo the market distress, the Fed will not meaningfully alter the course of policy.

I think there are two keys to understanding this lesson. First is Paul Samuelson’s quip that the market has called 9 out of the last 5 recessions.[1] Second is the fact, often repeated in the run up to liftoff, that the precise timing of any given policy action is of almost no importance.

Samuelson’s wisecrack is often taken to imply that the market gives highly unreliable signals about the state of the economy. But you don’t have to be Nate Silver to realize that calling 9 out of 5 recessions would indicate impressive predictive value. Taking Samuelson’s statement literally, more than half the time when the market signals recession, a recession ensues. In reality, the stock market probably is not as good a predictor as suggested by Samuelson, but stock market drops nonetheless reliably signal a heightened risk of recession.[2]

Given all the false signals, a market drop probably should not change ones view of the modal—that is, most likely—scenario much, but such drops can still suggest a significant rise in the probability of less likely, but far darker, scenarios.

Here is where the second point comes in: if the Fed has in place a policy trajectory that is appropriate for the modal scenario, there is essentially no cost briefly pausing the implementation of that trajectory to reassess macro fundamentals in light of a worrisome market signal. The effect of policy on the macroeconomy depends mainly on the expected course of policy over the next several quarters and years. Shifting the timing of any given policy move forward or backward a couple of FOMC meetings will have implications so small we cannot hope to measure them.

It is also important to note that no variable, not even the stock market, sends a reliable signal about recessions very far in advance—the stock market signal tends to be roughly contemporaneous. Thus, the Fed can get a reading on the validity of the stock market signal fairly quickly. If the stock market signal turns out to be a false alarm, the Fed can return to the previous course. The pause will have essentially no implications for the macroeconomy.[3]

In the world I’m discussing, the overall stance of policy is determined by macro fundamentals, with financial market signals occasionally putting wrinkles in the near-term timing. Chair Yellen’s explanation for the failure to liftoff at the September meeting squares with this view:

[I]n light of the heightened uncertainties abroad and a slightly softer expected path for inflation, the Committee judged it appropriate to wait for more evidence, including some further improvement in the labor market, to bolster its confidence that inflation will rise to 2 percent in the medium term. Now, I do not want to overplay the implications of these recent developments, which have not fundamentally altered our outlook. The economy has been performing well, and we expect it to continue to do so. [cite]

Here Yellen reiterates statements made by several policymakers before the September FOMC meeting that the U.S. outlook had not significantly changed. Some analysts take this fact as implying that policy should be unaffected, which then leads to the logic that the policy delay was a loss of nerve. Edward Luce put it this way just after Yellen’s September press conference:

So that is cleared up then. The Federal Reserve wanted to raise rates in September but then lost its nerve over China’s stock market crisis. Instead, it will probably move in December. No harm done.

The return to normal is on course barring a minor hiccup in the schedule. That, at least, was the gist of Janet Yellen’s message. Yet she also hinted she could simply repeat last week’s line in December.[cite]

Under this logic, the “gist of Janet Yellen’s message” must have amounted to Yellen making a tentative forecast of greater FOMC courage in December. My account (for better or worse) depends less on serial mood switching of the FOMC.

The modal outlook had not changed much, but the Fed was altering the near-term timing of liftoff—a move of little consequence—in order to reassess macro fundamentals in light of the market signal. If the market signal was not followed by the bottom falling out of the macro data, the previously signaled path would be resumed.

So what does this mean for policy in coming months? As in the weeks before September’s FOMC, policymakers have been out noting that near-term volatility has not greatly altered the modal outlook. It basically never does, which is Samuelson’s point. But pausing to assess whether macro fundamentals are turning seems even more clearly warranted than at the time of last September’s meeting.[4]

Let me add two forecasts. First, after the FOMC statement is released tomorrow, commentators will stumble over themselves in a race to divine whether hawks or doves have taken over the FOMC and will guide policy going forward. In our view, the FOMC statement and discussion thereafter is unlikely to imply any fundamental change, instead it may signal a pause to refresh the view on the macro fundamantals. If things quiet down and the macro economy continues to chug along, policy will also chug along, fundamentally unaltered, after a brief delay.

Second, as Bob Barbera and I have emphasized in a series of recent posts, we are not so sanguine about things chugging along. Instead, we see a darker picture for China and a weaker inflation outlook than seem consistent with the FOMC policy projections. If something like our forecast comes to pass, any pause will evolve into a much slower pace of rate increases than are implied by the FOMC’s December rate projections. Information arriving since we wrote this in December has nudged us further in the same direction.


1. “Science and Stocks,” Newsweek, September 19, 1966, p. 92 [back]

2. Much research supports the basic facts about the stock market lying behind Samuelson’s point. A nice recent piece on this is Bluedorn, et al., Do Asset Price Drops Foreshadow Recessions? IMF working paper 13/203. [back]

3. In this post, I am not arguing that the pause-tactic is optimal in any sense. I am arguing that it is a reasonable tactic and is how the FOMC seems to be behaving. For readers interested in formal optimality, I think it is clear that if properly communicated the conventional costs and or benefits of pausing would be second order. If I were attempting to build an interesting optimality argument, I’d include nonlinearities in the market signal, inherent nonlinearities in recessions which are magnified at the zero bound. Additionally, I’d take a world of imperfect information in which a pause might provide for more effective signal of conditional support for the economy in the event of recession than, say, would a simple Fed announcement conveying conditional support. [back]

4. The market signal is stronger and there are more negative hints showing in the U.S. macro data. [back]